Establishment of Monoclonal Anti-Retroviral gp70 ... - Journal of Virology

JOURNAL OF VIROLOGY, May 2000, p. 4116–4126 0022-538X/00/$04.00⫹0 Copyright © 2000, American Society for Microbiology. All Rights Reserved.

Vol. 74, No. 9

Establishment of Monoclonal Anti-Retroviral gp70 Autoantibodies from MRL/lpr Lupus Mice and Induction of Glomerular gp70 Deposition and Pathology by Transfer into Non-Autoimmune Mice NOBUTADA TABATA,1 MASAAKI MIYAZAWA,1,2* RYUICHI FUJISAWA,2† YUMIKO A. TAKEI,2 HIROYUKI ABE,1 AND KEIJI HASHIMOTO1,3 Department of Immunology1 and Third Department of Internal Medicine,3 Kinki University School of Medicine, Osaka-Sayama, Osaka 589-8511, and Department of Pathology, Tohoku University School of Medicine, Sendai 980-8575,2 Japan Received 6 October 1999/Accepted 1 February 2000

Several strains of mice, including MRL/MpJ mice homozygous for the Fas mutant lpr gene (MRL/lpr mice), F1 hybrids of New Zealand Black and New Zealand White mice, and BXSB/MpJ mice carrying a Y-linked autoimmune acceleration gene, spontaneously develop immune complex-mediated glomerulonephritis. The involvement of the envelope glycoprotein gp70 of an endogenous xenotropic virus in the formation of circulating immune complexes and their deposition in the glomerular lesions have been demonstrated, as has the pathogenicity of various antinuclear, antiphospholipid, and rheumatoid factor autoantibodies. In recent genetic linkage studies as well as in a study of cytokine-induced protection against nephritis development, the strongest association of serum levels of gp70–anti-gp70 immune complexes, rather than the levels of antinuclear autoantibodies, with the development and severity of glomerulonephritis has been demonstrated, suggesting a major pathogenic role of anti-gp70 autoantibodies in the lupus-prone mice. However, the pathogenicity of anti-gp70 autoantibodies has not yet been directly tested. To examine if anti-gp70 autoantibodies induce glomerular pathology, we established from unmanipulated MRL/lpr mice hybridoma clones that secrete monoclonal antibodies reactive with endogenous xenotropic viral env gene products. Upon transplantation, a high proportion of these anti-gp70 antibody-producing hybridoma clones induced in syngeneic non-autoimmune and severe combined immunodeficiency mice proliferative or wire loop-like glomerular lesions. Furthermore, deposition of gp70 in glomeruli and pathological changes were observed after intravenous injection of representative clones of purified anti-gp70 immunoglobulin G, demonstrating pathogenicity of at least some anti-gp70 autoantibodies. relationship between types of autoantibodies and the development of renal pathology. Several different clones of anti-DNA antibodies have been shown to induce glomerular lesions associated with immunoglobulin (Ig) deposition and/or proteinuria when transferred into non-autoimmune mice (13, 20, 36, 38). On the other hand, recent genetic analyses using simple sequence length polymorphisms as positional markers have identified several chromosomal loci in linkage with the development and severity of the renal disease. Interestingly, one of the genetic linkage analyses performed by using (NZB ⫻ NZW) F1 ⫻ NZW backcross mice (40) demonstrated that the loci linked with anti-gp70 Ab production, rather than those associated with levels of antinuclear Ab, had the strongest influence on the development of glomerulonephritis. In a similar study performed with C57BL/6 ⫻ (NZW ⫻ C57BL/6.Yaa)F1 backcross mice (26), association of serum levels of gp70 IC with severe glomerulonephritis was much stronger than that between levels of IgG anti-DNA autoantibodies and the renal disease. In addition, transgenic expression of interleukin-4 (IL-4) in the (NZW ⫻ C57BL/6.Yaa)F1 mouse model of SLE resulted in almost complete protection against the development of lupus-like nephritis in association with the lack of IgG3 production and marked decrease in the amount of serum gp70–anti-gp70 IC, while the serum concentrations of antiDNA IgG were not markedly reduced (25). These data suggest that autoantibodies reactive to endogenous retroviral gp70 comprise the major pathogenic Abs in the mouse models of

Several strains of mice such as MRL/MpJ mice homozygous for the Fas mutant lpr gene (MRL/lpr mice), F1 hybrids of New Zealand Black (NZB) and New Zealand White (NZW) mice [(NZB ⫻ NZW)F1], and BXSB/MpJ mice carrying a yet undefined Y-chromosome-associated autoimmune acceleration gene (Yaa) spontaneously develop an autoimmune syndrome closely resembling human systemic lupus erythematosus (SLE) (1, 4, 35, 41). Both human and murine SLE are serologically characterized by elevated levels of multiple autoantibodies (1, 4, 20, 35). These include antibodies (Abs) reactive with DNA and other nuclear components, Abs to extracellular matrices and cytoplasmic proteins, and in mice Abs reacting to the major envelope glycoprotein (gp70) of an endogenous xenotropic retrovirus that is expressed as a normal constituent of mouse serum (8). These autoantibodies and resultant circulating immune complexes (IC) have been implicated in the development of fatal glomerulonephritis. However, not all autoantibodies are primary pathogens (20, 39); in some cases they may instead be a secondary consequence of tissue damage. Several different approaches have been used to delineate the

* Corresponding author. Mailing address: Department of Immunology, Kinki University School of Medicine, 377-2 Ohno-Higashi, OsakaSayama, Osaka 589-8511, Japan. Phone and fax: 81 723-67-7660. E-mail: [email protected]. † Present address: Department of Microbiology, Kinki University School of Medicine, Osaka-Sayama, Osaka 589-8511, Japan. 4116

VOL. 74, 2000

MONOCLONAL ANTI-gp70 AUTOANTIBODIES FROM MRL/lpr MICE

4117

FIG. 1. Diagrammatic representation of the liver-derived gp70 cDNA and viral env genes and their chimeras expressed in recombinant vaccinia viruses. The nucleotide sequence of the gp70 cDNA isolated from an LPS-injected NZB mouse liver (29) is 99% homologous to that of the infectious NZB xenotropic virus env gene (21), except for several base changes clustered near the gp70/p15E cleavage site and in the 3⬘ flanking region and LTR (□), which are reflected by the indicated differences in restriction sites. The SFFV env gene ( ) is a product of natural recombination between endogenous polytropic and exogenous Friend ecotropic viruses with a large in-frame deletion (ƒ) encompassing the 3⬘ portion of gp70- and the 5⬘ portion of p15E-encoding regions. Dashed lines represent vector-derived sequences. A, AccI; B, BamHI; E, EcoRI; Ha, HaeII; Hc, HincII; Hd, HindIII; K, KpnI; S, SmaI; V, EcoRV; X, BstXI; AAAAA, poly(A) tail.

,

lupus nephritis. However, suggested pathogenicity of anti-gp70 autoantibodies has not yet been directly proven. Therefore, we decided to develop a new screening system and establish from MRL/lpr mice hybridoma clones that secrete monoclonal Abs (MAbs) reactive with endogenous xenotropic virus env gene products. MRL mice were chosen so that passive transfer into syngeneic mice of hybridoma cells and MAbs were more easily performed than in the cases of the F1 hybrid models with a complex genetic background. Tryptic peptide mapping analyses of gp70 molecules eluted from IC revealed that the serum gp70 involved in the production of circulating IC both in (NZB ⫻ NZW)F1 and MRL/lpr mice is structurally related to the envelope glycoprotein of an infectious NZB xenotropic virus (5, 12). Subsequent studies have shown that almost all strains of mice, healthy and SLE prone, produce endogenous xenotropic viral gp70 in the liver as an invariable serum constituent, and its expression is controlled as an acute-phase reactant (8). A cDNA clone encoding the serum gp70 was isolated from the liver of a lipopolysaccharide (LPS)-injected NZB mouse, and Northern blot analyses confirmed the expression of this message as an acute-phase reactant (29). Therefore, we used this cDNA clone, along with the env gene from an infectious molecular clone of NZB xenotropic virus (21), for in vitro expression of the endogenous retroviral env gene products to screen anti-gp70 Ab-producing hybridoma cells. Resultant hybridoma clones established from unmanipulated MRL/lpr mice induced severe glomerular lesions upon transplantation into syngeneic (BALB/c ⫻ MRL)F1 and severe combined immunodeficiency (SCID) mice. Moreover, purified IgG molecules of representative anti-gp70 autoantibodies induced glomerular deposition of gp70 and renal

pathology when injected intravenously (i.v.) into non-autoimmune mice. MATERIALS AND METHODS Mice. The original breeding pairs of MRL/MpJ-⫹/⫹ (MRL/⫹) and MRL/lpr mice were purchased from The Jackson Laboratory, Bar Harbor, Maine. These strains of mice were maintained by sister-brother mating in our animal facilities under specific-pathogen-free conditions. BALB/cCrSlc, NZW/NSlc, and C57BL/ 6CrSlc (B6) mice were purchased from Japan SLC, Inc., Hamamatsu, Japan, and (BALB/c ⫻ MRL/⫹)F1 hybrid mice were bred in our animal facilities. C.B-17/ Icr-scid/scid (SCID) mice were produced from the breeding pairs originally donated by S. Ikehara, Kansai Medical University, Moriguchi, Japan, and were kindly provided by M. Nose, Tohoku University School of Medicine. All animal experiments described in this report were approved by the institutions and performed under the guidelines of our animal facilities. NZB xenotropic virus-producing cells. NZB-AR cells that are chronically infected with a biological clone of NZB xenotropic virus were kindly provided by L. Evans, Laboratory of Persistent Viral Diseases, National Institute of Allergy and Infectious Diseases, Hamilton, Mont. Control uninfected Mv1Lu mink lung cells were purchased from the American Type Culture Collection, Manassas, Va. Expression of xenotropic murine leukemia viral env genes and their chimeras in recombinant vaccinia viruses. Vaccinia virus transfer vectors used for the expression of mouse retrovirus env genes and their chimeras were constructed as described previously (10, 17, 18). The structures of the expressed env genes and their chimeras are diagrammatically presented in Fig. 1. Plasmid clones pGP6-8, containing the gp70 cDNA isolated from a LPS-injected NZB mouse liver (29), and pNZB9-1, containing the whole permuted infectious molecular clone of an NZB xenotropic virus, IU-6 (21), were used as sources of endogenous xenotropic virus env gene sequences. Amino acid sequence analyses have revealed only three substitutions near the C terminus of gp70 between these two env gene products, although the C terminus of the transmembrane portion (p15E) contains five additional substitutions (21, 29), four of which are located within the R peptide that is cleaved from the mature transmembrane protein (31). A SalI oligonucleotide linker (New England Biolabs, Beverly, Mass.) was ligated onto both ends of the 2.2-kb AccI-HaeII fragment harboring the entire env sequence and a part of the long terminal repeat (LTR) isolated from pGP6-8 (Fig. 1), and the modified env-containing fragment was recloned into the unique SalI site of the

4118

TABATA ET AL.

vaccinia virus expression vector pSC11-SS (10). This construct was used to generate a vaccinia virus-NZB liver cDNA recombinant. The HincII-SmaI fragment harboring the entire env gene and portions of the pol and LTR from pNZB9-1 was reconstructed in pBluescript-KS(⫹) vector from purified HincIIEcoRI and EcoRI-SmaI fragments (Fig. 1), and the unique AccI site was replaced with a BamHI linker (New England Biolabs). A BamHI-digested fragment containing the entire env gene and a part of the LTR was cloned into the unique BglII site of the previously described modified vaccinia virus expression vector pSC11-SB (17), resulting in the generation of a vaccinia virus-infectious NZB xenotropic virus env gene recombinant. The env clones derived from the infectious NZB xenotropic virus and those derived from the NZB liver gp70 cDNA were easily distinguishable by the presence of a few different restriction sites (Fig. 1). For the construction of env gene chimeras, a plasmid clone (BT4-1a3 [42]) that contains a permuted infectious molecular clone of the Friend spleen focusforming virus (SFFV) was used as a source of nonxenotropic env sequences (Fig. 1). A vaccinia virus recombinant expressing the whole SFFV env gene has been described elsewhere (18). The EcoRI-SmaI fragment from pNZB9-1 was subcloned into pBluescript-KS(⫹) and was ligated with the 1.3-kb HindIII-EcoRI fragment harboring the 3⬘ portion of the pol and the 5⬘ portion of the SFFV env genes from BT4-1a3. The BamHI-digested fragment containing the entire chimeric env gene was then inserted to pSC11-SB at the unique BglII site. The resulting construct was used to generate a recombinant vaccinia virus that expressed the chimeric SFFV-NZB xenotropic virus env gene. For the construction of a reciprocal chimera, the unique KpnI site in the LTR of BT4-1a3 was replaced with the BamHI linker, and the 1.8-kb BamHI-digested fragment harboring the entire SFFV env gene was subcloned into pUC19. The HincII-EcoRI fragment containing the 5⬘ portion of the infectious NZB xenotropic virus env gene was ligated to the EcoRI-KpnI (BamHI) fragment of the subcloned SFFV env gene, taking advantage of the unique HincII site in the vector, and the AccI site upstream of the initiation site of NZB xenotropic virus env gene was replaced with a BamHI linker to insert the BamHI-digested fragment containing the chimeric env gene into pSB11-SB. The resultant plasmid was used to generate a recombinant vaccinia virus that expressed the chimeric NZB xenotropic virusSFFV env gene. Recombinant vaccinia viruses were produced by homologous recombination as described elsewhere (10, 17, 18). A recombinant vaccinia virus expressing the influenza virus hemagglutinin (HA) gene (30) was used as a negative control throughout the experiment. Production and screening of hybridoma cells. Spleen and lymph node cells were prepared aseptically from unmanipulated MRL/lpr mice. P3/NSI/1-Ag4-1 (NS-1) and P3X63Ag8.653 (8.653) myeloma cells were purchased from the American Type Culture Collection and used as fusion partner cells. Hybridoma cell fusion, hypoxanthine-aminopterin-thymidine selection, and cloning by colony formation in fibrin gels were performed as described previously (16, 24). For immunofluorescence detection of the reactivities of hybridoma Abs to expressed env gene products, monkey CV-1 cells were grown in wells of 96-well tissue culture plates, infected with a recombinant vaccinia virus at 100 to 200 PFU/well for 20 to 36 h, and incubated at 4°C overnight with a hybridoma culture supernatant added at 100 ␮l/well. Culture supernatants were then aspirated, and the wells were washed twice with phosphate-buffered balanced salt solution (PBBS) (3) containing 2% fetal bovine serum (FCS), and once with PBBS not containing FCS. Cells in each well were fixed with methanol, blocked with 10% skim milk, and stained with a 1/150 dilution of fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse Ig Ab (Cappel, Organon Teknika Corporation, West Chester, Pa.) as described elsewhere (17). For observation, the plates were placed upside-down under a Zeiss Axioplan fluorescence microscope (Zeiss, Overkochen, Germany). Antinuclear Ab activity was detected by treating uninfected CV-1 cells with methanol before incubating them with hybridoma-derived Ab. Methanol-fixed cells in wells of 96-well plates were washed with phosphatebuffered saline (PBS), blocked with 10% skim milk, and incubated with Ab as described above. To prepare representative immunofluorescence photographs, CV-1 cells were grown on glass coverslips and processed similarly. Hybridoma cells producing reference MAbs that react with various mouse retrovirus env gene products (2, 22, 23) were kindly provided by B. Chesebro, Laboratory of Persistent Viral Diseases, National Institute of Allergy and Infectious Diseases. Hybridoma cell line N-S.7, producing mouse IgG3 reacting with sheep red blood cells (SRBC), was purchased from the American Type Culture Collection. Another IgG3-producing hybridoma clone, 11, reactive with mumps virus nucleoprotein (37), was kindly provided by Y. Ito, Department of Microbiology, Mie University School of Medicine, Tsu, Japan. Ig isotypes of MAbs were determined by an Ouchterlony immunodiffusion method using an isotypespecific Ab kit (The Binding Site, Birmingham, United Kingdom) as described previously (16, 24). The above-described reference MAbs and others of previously defined isotypes were used as controls in the Ig isotype determination. Western blotting. Western blotting analyses of polypeptide specificity of the Abs was performed as described previously (17, 19, 22, 23), using extracts from NZB-AR and control Mv1Lu cells. In brief, cells were washed four times with ice-cold PBS and incubated with 0.5% NP-40 in 50 mM Tris-buffered saline (pH 7.4) containing 10 mM EDTA, 5 mM n-ethylmaleimide, 1 mM phenylmethylsulfonyl fluoride, and 0.002% leupeptin at 4°C for 15 min. The supernatant was collected after centrifugation at 15,000 ⫻ g for 10 min. The extract was mixed with an equal volume of 4% sodium dodecyl sulfate (SDS) sample buffer (17, 19)

J. VIROL. without a reducing agent and was subjected to SDS-polyacrylamide gel electrophoresis. Proteins separated through 7.5% polyacrylamide gels were transferred onto polyvinylidene difluoride membranes (Immobilon; Millipore Corporation, Bedford, Mass.) as described previously (17, 19), and the blotted membrane was blocked with 10% skim milk. Incubation with MAb and detection of bound Ab by using biotinylated horse anti-mouse Ig secondary Ab and avidin-biotinylated peroxidase complex (Vector Laboratories, Burlingame, CA) has been described elsewhere (17, 19). For the detection of serum gp70, sera from NZW, (BALB/c ⫻ MRL/⫹)F1, and B6 mice were mixed at 1:20 with the SDS sample buffer containing no reducing agent, and serum proteins were separated through 7.5% polyacrylamide gels and blotted as described above. Serum gp70 molecules were detected with biotinconjugated anti-gp70 MAb 24-6 by chemiluminescence reaction using horseradish peroxidase-conjugated streptavidin (Vector Laboratories) and ECL⫹ reagent (Amersham Pharmacia Biotech, Uppsala, Sweden) according to the manufacturers’ instructions. Transfer of hybridoma cells or purified Abs into mice and pathological analyses. Hybridoma cells were grown in Dulbecco’s modified Eagle medium supplemented with glucose (4.5 g/liter [final concentration]), gentamicin sulfate (50 mg/liter), and 10% FCS, washed twice with PBBS, and resuspended in PBBS at 107 cells/ml. (BALB/c ⫻ MRL/⫹)F1 and SCID mice were transplanted intraperitoneally (i.p.) with 1 ⫻ 107 to 2 ⫻ 107 hybridoma cells after a pretreatment with a 0.5-ml/mouse i.p. dose of 2,6,10,14-tetramethylpentadecane (pristane; Aldrich Chemical Co., Inc., Tokyo, Japan) given 1 to 3 weeks prior to hybridoma transplantation. Serum concentrations of IgG in transplanted SCID mice were measured by single radial immunodiffusion assays using isotype-specific antisera (anti-mouse IgG2a and anti-mouse IgG3; Zymed Laboratories, Inc., South San Francisco, Calif.) as described previously (34). For purification of a clonal antigp70 IgG, hybridoma cells were grown in a serum-free medium (Hybridoma SFM; Gibco BRL, Rockville, Md.) in 4-liter spinner flasks, and culture supernatants were concentrated by using a tangential flow ultrafiltration system (Minitan II; Millipore Corporation). IgG was purified by protein A-Sepharose (Amersham Pharmacia Biotech) affinity chromatography as described previously (19, 24). Special care was taken to perform the purification aseptically at room temperature. Purified MAbs dissolved in PBBS at 0.5 to 1.0 mg/ml were injected into the tail vein after removing possibly contaminating Ig aggregates by centrifugation at 10,000 ⫻ g for 15 min. The methods of preparation and staining of formalin-fixed, paraffin-embedded tissue sections and specimens for electron microscopy have been described elsewhere (19, 34). A part of the kidneys from each mouse was snap frozen in a mixture of dry ice and acetone after being embedded in O.C.T. compound (Miles Scientific, Naperville, Ind.), and frozen sections were prepared as described previously (16, 19, 24). For immunofluorescence detection of mouse IgG and C3 in frozen sections, FITC-conjugated goat anti-mouse IgG and anti-mouse C3 Ab (Cappel, Organon Teknika Corporation) were used. To detect the deposition of retroviral gp70, MAbs specific for xenotropic viral env gene products, 24-6 and 24-9 (23), were purified as described above and labeled with biotin (19, 24). Localization of the biotinylated anti-xenotropic viral envelope MAb was visualized by using the avidin-biotinylated peroxidase complex (Vector Laboratories) as described previously (16, 24). Histopathologic severity of each glomerular lesion in periodic acid-Schiff (PAS)-stained sections was semiquantitatively determined according to previously described criteria (34), and an average index of glomerular pathology (IGP) was calculated by examining ⬎20 glomeruli per mouse. In brief, grade 1 was given when there was apparent increase in the number of mesangial cells (⬎3 nuclei in a single separate section of a mesangial area) but no inflammatory cell infiltration into capillaries, grade 2 was given when cellular components were increased in at least one capillary lumen, and grade 3 was given when obliteration of at least one capillary lumen with fibrin- or collagen-containing materials was observed. Grade 0 means that none of the above histologic changes were observed in a glomerulus in question. Mice in which ⬎80% of examined glomeruli showed significant histologic changes (IGP ⱖ 1) or in which 30 to 80% of glomeruli showed severe histologic changes (IGP ⱖ 2) are designated nephritic in this study. Incidences and average IGP were statistically compared with those of control mice by Fisher’s exact probability test and by Student’s t test, respectively.

RESULTS Expression of endogenous xenotropic viral env cDNA and establishment of MAbs reactive with the env gene product from MRL/lpr mice. A DNA fragment containing the entire env gene sequence from the cDNA clone isolated from a LPSinjected NZB mouse was inserted into a vaccinia virus expression vector, and a recombinant vaccinia virus that expressed the liver-derived xenotropic viral envelope glycoprotein was constructed (Fig. 1). The whole env gene from a molecular clone of an infectious xenotropic virus isolated from an NZB mouse, IU-6, was also expressed in another vaccinia virus re-

VOL. 74, 2000


4119

FIG. 2. Antigenic characterization of expressed env gene products using a panel of MAbs. Representative immunofluorescence micrographs of infected CV-1 cells are presented. 24-6, 24-8, 514, and 603 are anti-retroviral envelope MAbs of previously defined virus type and polypeptide specificities (2, 22, 23), whose reported reactivities to different types of mouse C-type retroviruses are given in parentheses; 12H5.1 is a representative MAb newly established from MRL/lpr mice in this study. Note that only the foci of cells infected with a relevant recombinant vaccinia virus, not the uninfected cells surrounding the foci, are stained.

combinant as a positive control. Reactivities of a panel of antiretroviral MAbs previously established from non-autoimmune mice (2, 22, 23) to these two xenotropic viral env gene products showed no difference (Fig. 2 and Table 1), reflecting their almost identical amino acid sequences. Fifteen separate hybridoma clones were selected from a total of three fusions of spleen and lymph node cells, using eight unmanipulated, female MRL/lpr mice, both for reactivity of secreted Ab with CV-1 cells expressing the liver-derived gp70 cDNA and for lack of reactivity to cells infected with the control vaccinia virus-influenza virus HA recombinant, as exemplified in Fig. 2. Of these, three were established from the first fusion performed by using spleen and lymph node cells from two 2.5-month-old female mice and 8.653 myeloma cells, seven others were established from the second fusion in which four 2.5 month-old female mice and NS-1 myeloma cells were used, and the remaining five clones were derived from the third fusion performed by using two 4.5-month-old female mice and NS-1. An additional clone, 17D7.1, was similarly selected from a fusion made with spleen and lymph node cells of two 4.5month-old male MRL/lpr mice and 8.653 myeloma cells. A few nonproducer clones were also established from these fusions, as represented by clone 4E9.1 in Table 2, and were used as negative controls in the following experiment along with the fusion partner cells. During the initial screening procedure for Ab-producing hybridoma cells, wells containing antinuclear Ab

were also observed at roughly the same frequency as those containing anti-gp70 Ab; however, none of the MAbs selected for reactivity to the xenotropic viral env gene product crossreacted with nuclear antigens in the immunofluorescence assay. Western blotting analysis confirmed the reactivity of these MAbs with the whole env gene product gp85 (gp70 plus p15E), which was detected from the lysate of NZB-AR cells chronically infected with an NZB xenotropic virus but not from the lysate of uninfected Mv1Lu cells (Fig. 3). Although Abs reactive with the surface components of CV-1 cells other than expressed gp70 were eliminated through the screening procedure, by selecting MAbs reactive with the plaques of gp70expressing cells but not with the surrounding uninfected CV-1 cells (Fig. 2), a few bands other than that of gp85 were readily detectable with some of these MAbs in blots of both NZB-AR and uninfected Mv1Lu cell lysates, suggesting possible crossreactivity with normal cellular components. The 16 MAbs reactive with the xenotropic viral env gene products were further analyzed for reactivities to the products of the SFFV env gene and the chimeras between NZB xenotropic virus and SFFV env genes (Fig. 1 and Table 2) for rough epitope mapping. The SFFV env gene is a naturally produced recombinant with a large deletion encompassing the C-terminal one-fourth of the gp70 and the N-terminal half of the transmembrane p15E (42). Its gp70 sequence is unrelated to that of Friend murine leukemia virus, but the N-terminal one-

TABLE 1. Reactivities of reference MAbs of known specificity to various murine leukemia virus env gene products Reactivity with CV-1 cells expressingb:

Hybridoma clone

Virus type, polypeptide specificitya

NZB liver gp70 cDNA

NZB IU-6 env

SFFV env

NZB ⫹ SFFV env

SFFV ⫹ NZB env

Influenza virus HA

24-6 24-8 514 603 11

Xeno ⫹ Poly, gp70 Eco ⫹ Poly, gp85 Poly, gp70 Xeno, gp70 Mumps, NP

⫹ ⫺ ⫺ ⫹ ⫺

⫹ ⫺ ⫺ ⫹ ⫺

⫺ ⫺ ⫹ ⫹ ⫺

⫹ ⫺ ⫺ ⫹ ⫺

⫺ ⫺ ⫹ ⫹ ⫺

⫺ ⫺ ⫺ ⫺ ⫺

a Virus type specificities of reference MAbs are summarized from the previous reports (2, 22, 23). Xeno, xenotropic viruses; Poly, polytropic viruses; Eco, ecotropic viruses; NP, nucleoprotein. b Tested by indirect immunofluorescence assays using CV-1 cells infected with an indicated recombinant vaccinia virus as target cells as shown in Fig. 2. Assays were repeated at least three times for each MAb, with consistent results.

4120

TABATA ET AL.

J. VIROL.

TABLE 2. Characteristics of anti-xenotropic viral MAbs established from MRL/lpr mice and incidence and severity of glomerular lesions induced by transplantation of the hybridoma cellsa Hybridoma clone

Heavy-chain isotype

8.653 NS-1 4E9.1b (nonproducer) 17D7.1 34B4.1 58C5.1 12H5.1 36D1.1 37C6.1 42B4.1 42D3.2 6F12.3 7C6.3 51C4.1 51D1.1 59C4.1 60A5.1 37C4.1 42D3.1

(⫺) (⫺) (⫺) ␮ ␮ ␮ ␥3 ␥2a ␥2a ␥2a ␥2a ␮ ␮ ␥3 ␥3 ␥3 ␥3 ␥3 ␥2a

Reactivity with CV-1 cells expressing:

Glomerular pathology

NZB liver gp70 cDNA

NZB IU-6 env

SFFV env

NZB ⫹ SFFV env

SFFV ⫹ NZB env

Influenza virus HA

Incidence

Avg IGP ⫾ SEM

⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫺

⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫺

⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺

1/7 0/7 0/6 0/17 0/8 10/12c 12/13c 0/11 10/14c 0/8 1/7 0/12 1/7 1/4 14/19c 4/8 6/9d 12/12c 0/10

0.42 ⫾ 0.10 0.12 ⫾ 0.03 0.22 ⫾ 0.12 0.52 ⫾ 0.05 0.34 ⫾ 0.08 1.07 ⫾ 0.08e 2.28 ⫾ 0.18e 0.23 ⫾ 0.05 0.96 ⫾ 0.06e 0.49 ⫾ 0.05e 0.54 ⫾ 0.11f 0.30 ⫾ 0.06 0.49 ⫾ 0.09 0.75 ⫾ 0.15f 1.20 ⫾ 0.12e 0.88 ⫾ 0.16f 1.01 ⫾ 0.09e 2.10 ⫾ 0.19e 0.49 ⫾ 0.06

a Reactivities of MAbs were tested by indirect immunofluorescence assays as described for Table 1. Criteria for histopathologic diagnosis of glomerulonephritis and the standards for determination of the IGP are described in Materials and Methods. The incidence of nephritis and IGP for each MAb were statistically compared with those for relevant fusion partner cells. b Three other nonproducer or non-anti-gp70 hybridoma clones were similarly tested for potential pathogenicity, and none induced significant glomerular pathology in transplanted (BALB/c ⫻ MRL/⫹)F1 mice. c P ⬍ 0.003 by Fisher’s exact probability test. d 0.01 ⬍ P ⬍ 0.02 by Fisher’s exact probability test. e P ⬍ 0.001 by Student’s t test. f 0.001 ⬍ P ⬍ 0.01 by the t test.

third is most similar to endogenous polytropic viruses (31, 42), a type distinct from xenotropic viruses. Eight of the MRL/lprderived MAbs (17D7.1 through 42D3.2 in Table 2) reacted with the products of the both xenotropic viral env genes but lost reactivity when the 5⬘ one-third of the NZB xenotropic viral env gene was replaced at the EcoRI site with the corresponding portion of SFFV env. Based on their reactivity to the reciprocal chimera (NZB xenotropic viral env-SFFV env), they are most likely to react with epitopes located in the N-terminal

one-third of xenotropic viral gp70. On the other hand, two clones, 37C4.1 and 42D3.1, reacted with the products of the whole NZB xenotropic viral env and the SFFV env-NZB xenotropic viral env chimera but not with the products of the SFFV env and the NZB xenotropic viral env-SFFV env chimera. Therefore, they seem to recognize epitopes located in the C-terminal portion of the xenotropic viral env gene products. Six other clones were reactive to the products of xenotropic viral and SFFV env genes and both of the chimeras.

FIG. 3. Representative results of Western blotting assays showing virus polypeptide specificities of the MAbs. Mr, molecular mass markers, with positions indicated in kilodaltons at the left; U, uninfected mink Mv1Lu cells; AR, NZB-AR cells chronically infected with a biological clone of NZB xenotropic virus. N-S.7 is a negative control IgG3 specific for SRBC; 603 is a positive control IgM specific for xenotropic viral gp70. The arrowhead indicates bands of the viral env gene product, gp85 (gp70 plus p15E).

VOL. 74, 2000


These MAbs, therefore, may recognize epitopes common to different types of mouse retrovirus env gene products. It is notable that these latter MAbs showed more prominent crossreactivity to normal cellular components in Western blotting as exemplified by clones 51D1.1 and 603 in Fig. 3. Pathogenicity of gp70-reactive autoantibodies produced from transplanted hybridoma cells in non-autoimmune mice. To test possible pathogenicity of these MAbs reactive with retroviral gp70, each hybridoma clone was injected i.p. into syngeneic (BALB/c ⫻ MRL/⫹)F1 mice that had been injected with a single i.p. dose of pristane to facilitate hybridoma transplantation. Injected mice were killed before dying of a tumor burden, and the organs were examined histopathologically. The average interval between pristane injection and organ removal was 23.8 days, and that between hybridoma transplantation and organ removal was 14.3 days. Six of the 16 hybridoma clones induced in syngeneic (BALB/c ⫻ MRL/⫹)F1 mice significant glomerular lesions at a considerable frequency, while the transplantation of fusion partner cells or a control nonproducer clone did not induce significant pathology at this early stage after a single pristane treatment (Fig. 4 and Table 2). Another clone, 59C4.1, induced histologically evident glomerular lesions (Fig. 4l) at a low frequency (in four of eight mice). Among the six clones that consistently induced glomerular lesions, four IgG3-producing hybridoma clones, 12H5.1, 37C4.1, 51D1.1, and 60A5.1, caused diffuse and histologically more severe glomerular pathology compared with other antigp70 hybridomas when transplanted into (BALB/c ⫻ MRL/ ⫹)F1 mice. The glomerular lesions induced by transplantation of hybridoma clone 12H5.1 were characterized by intracapillary proliferation and/or infiltration of cells with granular subendothelial and intracellular deposition of IgG and C3 (Fig. 4b to d). Massive deposition of fibrin was also demonstrated in affected glomeruli by phosphotungstenic acid-hematoxylin staining (not shown). On the other hand, hybridoma clone 37C4.1 induced diffuse lupus-like glomerular lesions characterized by light microscopic wire loops and massive subendothelial IgG deposition (Fig. 4g and h). Two other IgG3 clones, 51D1.1 and 60A5.1, induced proliferative glomerular lesions at a high incidence with dilatation of capillary lumina and occasional accumulation of red cell fragments (Fig. 4k). One clone (58C5.1) out of five IgM- and another (37C6.1) from five IgG2a-producing hybridoma cells also induced proliferative glomerular lesions at a high frequency (Table 2). Induction of glomerular lesions in transplanted SCID mice and deposition of gp70 in glomeruli. To exclude the possibility that the induction of glomerular lesions by transplantation of the hybridoma cells was due to host immune responses against hybridoma-derived Abs or cellular components, or a result of autoantibody production in response to pristane injection, we next injected three representative clones of the hybridoma cells into SCID mice. Both hybridomas 12H5.1 and 37C4.1 induced in the transplanted SCID mice glomerular lesions that were histologically similar to the lesions induced in the F1 mice (Fig. 4e, f, i, and j). The presence of xenotropic viral gp70, along with IgG and complement, was demonstrated in glomerular lesions of SCID mice transplanted with either one of the two pathogenic hybridoma clones (Fig. 4f and j), suggesting the deposition of gp70 IC. On the other hand, an IgG2a-producing anti-gp70 hybridoma clone, 36D1.1, did not induce significant nephritic lesions in SCID mice. Since differential measurement of the concentrations of IgG produced from transplanted hybridoma cells was impractical in immunocompetent (BALB/c ⫻ MRL/⫹)F1 mice, serum concentrations of hybridoma-derived IgG were determined by single radial immunodiffusion in SCID mice. At

4121

the time the transplanted SCID mice were killed for histopathologic examination, average serum concentration of IgG2a in hybridoma 36D1.1-bearing mice was 6.7 mg/ml, while that of IgG3 in hybridoma 12H5.1-bearing mice was 5.3 mg/ml. In some SCID mice transplanted with hybridoma 36D1.1 cells, higher serum concentrations of IgG2a such as 16.6 mg/ml were observed. Induction of gp70 deposition and glomerular pathology by injecting purified MAb. To further exclude the possibility that the glomerular lesions were induced by products of the hybridoma cells other than Ig, IgG3 molecules purified from culture supernatants of hybridoma cells 12H5.1 and 51D1.1 were injected i.v. into syngeneic (BALB/c ⫻ MRL/⫹)F1 mice. A single injection of purified 12H5.1 induced minimal glomerular pathology; however, when purified 12H5.1 IgG3 (0.25 mg/mouse) was injected for 3 consecutive days and the kidneys were examined 2 days after the final injection, diffuse granular deposition of retroviral gp70 in the glomeruli was observed by immunohistochemical staining (Fig. 5c and d) along with IgG. No gp70 deposition was observed when control anti-SRBC IgG3 was injected in the same manner (Fig. 5e). Histologic changes characterized by PAS-positive depositions in the mesangial area were also observed in all the mice injected with purified 12H5.1 IgG (Fig. 5g) but not in those injected with purified anti-SRBC IgG (Fig. 5f). Repeated injection of purified 51D1.1 on the same schedule resulted in slight expansion of mesangial areas and minimal deposition of gp70. However, when purified 12H5.1 and 51D1.1 were mixed and injected for 3 consecutive days as described above, apparently more severe glomerular pathology characterized by edema and PAS-positive deposits in the mesangial areas and some capillary walls was observed (Fig. 5h). Thus, these results directly indicate that at least some monoclonal anti-gp70 autoantibodies induce, in the absence of other cellular products, glomerular deposition of gp70 and renal pathology. Differences in glomerular pathology in mice expressing high and low levels of serum gp70. To further examine the possibility that injected anti-gp70 autoantibodies were involved in the formation of immune complexes with serum gp70, purified MAb 12H5.1 was injected i.v. into three different strains of mice that are known to express high or low levels of serum gp70. NZW mice have been shown to express the highest level of serum gp70 among several different strains tested (14), while B6 mice express a very low level of gp70 in their sera (15). These differences in the amount of expressed serum gp70 were also confirmed by Western blotting, along with the expression of a relatively large amount of serum gp70 in (BALB/c ⫻ MRL/⫹)F1 mice (Fig. 6a). Although NZW and B6 mice are not syngeneic to BALB/c and MRL backgrounds in which the hybridoma cells were produced, mouse IgG3 constant regions contain extremely limited polymorphisms (32), and no serologically definable IgG3 allotypes have been reported. Thus, induction of anti-allotypic immune responses by injecting purified IgG3 molecules is very unlikely. When IgG3 molecules purified from culture supernatants of 12H5.1 or control N.S-7 hybridoma cells were injected for 3 consecutive days as described above, all of the 10 NZW and 8 (BALB/c ⫻ MRL/⫹)F1 mice injected with 12H5.1 IgG3 developed focal but significant glomerular pathologies characterized by thickening of the capillary walls, cell proliferation, and inflammatory cell infiltration (Fig. 6b and c). On the other hand, significant pathologic changes were not observed in the B6 mice injected with purified 12H5.1 IgG3 (Fig. 6d). Control N.S-7 IgG3 did not induce significant glomerular lesions in any of the three strains of mice. Furthermore, no deposition of

4122

TABATA ET AL.

J. VIROL.

FIG. 4. Representative kidney pathology of hybridoma-transplanted mice. (a) (BALB/c ⫻ MRL/⫹)F1 mouse transplanted with 8.653 fusion partner cells. PAS staining, ⫻70. (b) (BALB/c ⫻ MRL/⫹)F1 mouse transplanted with hybridoma cells 12H5.1. PAS staining, ⫻70. Note the extreme expansion of the glomeruli compared to those in panel a, which shows normal glomeruli at the same magnification. Sizes of tubules and of the nuclei of tubular epithelial cells are not different in panels a and b, but glomeruli are markedly enlarged in panel b. Granular deposition of fibrin in the affected glomeruli was also shown when phosphotungstenic acidhematoxylin staining was applied (not shown). (c) Immunofluorescence staining with FITC-conjugated anti-mouse IgG of a fresh-frozen section taken from a representative (BALB/c ⫻ MRL/⫹)F1 mouse transplanted with hybridoma cells 12H5.1. Use of FITC-conjugated anti-mouse C3 resulted in a similar pattern of staining. (d) Electron micrograph showing an affected glomerulus of a representative (BALB/c ⫻ MRL/⫹)F1 mouse transplanted with hybridoma 12H5.1. Cells occupying the capillary lumina (❉) are filled with numerous electron-dense granules. Arrows indicate subendothelial deposits along the basement membrane. Bar ⫽ 2 ␮m. (e) SCID mouse transplanted with hybridoma cells 12H5.1. PAS staining, ⫻140. (f) Immunoperoxidase staining of a fresh-frozen section prepared from a SCID mouse at 3 days after transplantation of hybridoma 12H5.1. Purified MAb 24-9 (23) was biotinylated to detect the presence of xenotropic viral env gene products. (g) (BALB/c ⫻ MRL/⫹)F1 mouse transplanted with hybridoma cells 37C4.1 showing typical wire loop lesions. PAS staining, ⫻175. (h) Electron micrograph of the kidney from a (BALB/c ⫻ MRL/⫹)F1 mouse transplanted with hybridoma cells 37C4.1. Bar ⫽ 2 ␮m. Note the dense subendothelial deposits consistent with light microscopic wire loops along the basement membrane. (i) SCID mouse transplanted with hybridoma cells 37C4.1. PAS staining, ⫻140. (j) Dense linear deposition of mouse C3 in a representative glomerulus from a SCID mouse transplanted with hybridoma 37C4.1. Similar deposition of mouse IgG and xenotropic viral gp70 in the affected glomeruli was also demonstrated in fresh-frozen sections of the transplanted SCID mice. (k and l) (BALB/c ⫻ MRL/⫹)F1 mice transplanted with one clone of hybridomas 51D1.1 and 59C4.1, respectively. PAS staining, ⫻140. Note PAS-positive deposition and expansion of the mesangial areas in panel k and cell proliferation (arrowheads) and occlusive changes (arrow) of capillaries in panel l. Lesions similar to those in panel l were observed in the mice transplanted with hybridoma 60A5.1 (not shown).

gp70 was demonstrated in the kidneys of B6 mice after injection of purified 12H5.1 IgG3. DISCUSSION In this study, we established from unmanipulated MRL/lpr lupus mice hybridoma clones that secrete MAbs reactive with

the endogenous xenotropic viral env gene products. The MAbs were selected both for reactivity with CV-1 cells expressing the xenotropic viral env cDNA and for lack of reactivity with the same cells expressing the influenza virus HA gene. Specificities of the established MAbs were further confirmed by Western blotting and immunofluorescence assays using CV-1 cells expressing chimeric env genes between NZB xenotropic virus and

VOL. 74, 2000


4123

FIG. 5. Photomicrographs showing gp70 deposition in glomeruli and pathology induced in mice injected with purified anti-gp70 MAb. (a [⫻85] and b [⫻175]) Representative frozen sections taken from a female MRL/lpr mouse showing granular deposition of gp70 in glomeruli; immunoperoxidase staining with MAb 24-6. (c [⫻85] and d [⫻175]) Representative frozen sections taken from a (BALB/c ⫻ MRL/⫹)F1 mouse injected with purified 12H5.1 IgG3; immunoperoxidase staining with MAb 24-6. Note that all four glomeruli seen in panel c (arrows) exhibit gp70 deposition. Deposits of gp70 seem to localize along mesangial cells (d). (e) Representative frozen section taken from a control (BALB/c ⫻ MRL/⫹)F1 mouse injected with purified N-S.7 IgG3. No gp70 deposition was observed. (f) Representative glomeruli of a control (BALB/c ⫻ MRL/⫹)F1 mouse injected with purified N-S.7 IgG3; PAS staining, ⫻175. (g) Representative glomerular pathology induced by injection of purified 12H5.1 IgG3 in (BALB/c ⫻ MRL/⫹)F1 mice; PAS staining, ⫻175. Note expansion of mesangial areas with PAS-positive deposits (arrowhead). (h) Representative glomerular pathology induced in (BALB/c ⫻ MRL/⫹)F1 mice by injecting a mixture of purified 12H5.1 and 51D1.1 IgG3; PAS staining, ⫻350. Note expansion of mesangial spaces between the capillaries and subendothelial hyaline deposits (arrowhead).

Friend SFFV. About one-half of the gp70-reactive MAbs established from MRL/lpr mice lost reactivity when a part of the xenotropic viral env gene was replaced with the corresponding portion of SFFV env, thus confirming the presence of antigenic epitopes within the xenotropic viral gp70. On the other hand,

some other MAbs similarly established from MRL/lpr mice were reactive to both the xenotropic viral and SFFV env gene products. These latter MAbs, exemplified by clone 51D1.1, tended to show reactivities to several protein bands other than gp85 that were common to uninfected Mv1Lu mink cells and

4124

TABATA ET AL.

J. VIROL.

FIG. 6. Differences in serum gp70 expression and glomerular pathology induced after injection of purified 12H5.1 IgG3 in NZW, (BALB/c ⫻ MRL/⫹)F1, and B6 mice. (a) Results of Western blotting assays showing the expression of serum gp70 in three different strains of mice. Sera were diluted 1:20 into SDS sample buffer without a reducing reagent and boiled for 5 min; 10 ␮l of each boiled mixture was loaded into a well of 7.5% polyacrylamide gel. Plasma from a 4 month-old female MRL/lpr mouse (Lpr), which should contain a large amount of gp70–anti-gp70 immune complexes, was used as a positive control. As reported previously (14, 15), NZW mice expressed a high level of serum gp85 (gp70 plus p15E), gp70, and a degradation product gp45 (arrowheads), while their expression in B6 mice was low. (BALB/c ⫻ MRL/⫹)F1 mice (F1) expressed an intermediate level of serum gp70. Mr, biotinylated markers, with positions indicated in kilodaltons at the left. (b to d) Representative photomicrographs taken from kidney sections of NZW (b), (BALB/c ⫻ MRL/⫹)F1 (c), and B6 (d) mice injected with purified anti-gp70 IgG3, 12H5.1; hematoxylin and eosin staining, ⫻300. Note apparent thickening of the capillary walls (arrowheads) and inflammatory cell infiltration (arrow) in panel b and marked increase in glomerular cellularity and evident neutrophilic infiltration in panel c.

NZB-AR cells chronically infected with an NZB xenotropic virus in Western blotting. Possible cross-reactivity of these gp70-reactive antibodies with normal cellular components and its potential roles in the development of autoimmune lesions, described as molecular mimicry (6), might be worth pursuing. Glomerular lesions were induced in non-autoimmune mice by transplanting single clones of hybridoma cells producing gp70-reactive Abs. At least 7 of the 16 separate hybridoma clones established from unmanipulated MRL/lpr mice induced histopathologically significant glomerular lesions, and deposition of xenotropic viral gp70 along with IgG and C3 was demonstrated in the lesions induced by transplantation of the two representative hybridoma clones into SCID mice. Direct involvement of the anti-gp70 MAb, rather than possible secondary host immune responses to the transplanted hybridoma cells including anti-idiotypic Ab production, in the induction of glomerular pathology was demonstrated by successful induction in SCID mice of glomerular lesions that were similar to those induced in (BALB/c ⫻ MRL/⫹)F1 mice. It has been shown that a single i.p. injection of pristane induces in non-autoimmune BALB/c mice production of anti-

nuclear ribonucleoprotein and anti-Su autoantibody production and proliferative glomerulonephritis (27, 28). However, the development of autoantibody production and nephritis took months after pristane injection (27). On the other hand, our mice were killed and examined within 5 weeks after a single pristane injection, and thus it is unlikely that the injection of pristane alone was responsible for the development of glomerular lesions in the hybridoma-transplanted mice. In fact, control mice transplanted with the fusion partner cells or a nonproducer clone of hybridoma cells established from MRL/ lpr mice after an injection of the same pristane dose showed only minimal pathologic changes in the kidneys (Fig. 4 and Table 2). Reproduction of severe glomerular pathology in SCID mice that should not produce any Ab in response to pristane injection (Fig. 4) also supports the notion that pristane-induced autoantibody production is not the major pathogenetic factor in this transplantation model. It should be noted that possible production of anti-gp70 autoantibodies in the above-described pristane-induced model of nephritis has not been examined. Thus, the presence of the pristane-induced model neither contradicts nor supports possible pathogenicity

VOL. 74, 2000


of anti-gp70 autoantibodies. It is also possible, however, that production of some cytokines, especially IL-6, either from the hybridoma cells or from host tissues in response to pristane injection and/or hybridoma transplantation, might have contributed to the development of glomerular pathology. The slight increase in the index of glomerular pathology in mice transplanted with 8.653 myeloma cells (Table 2) might be explained by this mechanism. However, representative clones of the anti-gp70 MAb did induce glomerular deposition of gp70 and significant pathology in non-autoimmune mice when injected as purified IgG, clearly eliminating possible pathogenetic effects of cellular products other than Ig. Possible involvement of gp70–anti-gp70 immune complexes in the induction of the currently described Ab transfer models was indicated by the demonstration of gp70 deposition in affected glomeruli along with IgG and C3 (Fig. 4 and 5). This finding was supported by the demonstration of differences in glomerular pathology induced by injection of purified antigp70 MAb 12H5.1 in NZW and B6 strains of mice that are known to express high and low serum levels of gp70, respectively (Fig. 6). Thus, mice expressing high levels of serum gp70 developed apparently more severe glomerular pathology after injection of 12H5.1 IgG3, while B6 mice expressing minimal serum gp70 did not develop glomerular lesions even when the same amount of purified anti-gp70 IgG3 was injected. These results support the possibility that injected anti-gp70 MAb produced immune complexes with serum gp70 before being deposited into kidney glomeruli. Further studies including measurements of serum immune complexes are required to correlate possible production of gp70–anti-gp70 immune complexes and the development of glomerulonephritis. It is noteworthy that anti-gp70 IgG3-producing hybridoma clones induced histologically evident glomerular lesions at an apparently higher frequency than IgM- and IgG2a-producing clones did (Table 2). Since it is difficult to differentially measure the amount of IgG produced from transplanted hybridoma cells in immunocompetent (BALB/c ⫻ MRL/⫹)F1 mice, concentrations of hybridoma-derived IgG were determined for a limited number of hybridoma clones in transplanted SCID mice. Average concentrations of serum IgG were in the same range among the mice transplanted with a representative IgG2a-producing hybridoma cells and those bearing representative IgG3-producing cells. Although hybridoma-derived Ab concentrations were not measured in every transplanted animal, it is unlikely that IgM- and IgG2a-producing hybridoma cells, but not IgG3-producing ones, selectively lose their Abproducing ability soon after transplantation. Therefore, IgG3 anti-gp70 MAbs may have higher pathogenic potentials than IgM and IgG2a anti-gp70 MAbs. In fact, the importance of the IgG3 isotype in the induction of glomerular lesions has been demonstrated in spontaneous and induced models of MRL/lpr mice (7, 34). The importance of IgG3 isotype in the pathogenesis of mouse SLE was also demonstrated in a recent study (25) in which transgenic expression of IL-4 protected (NZW ⫻ C57BL/6.Yaa)F1 lupus mice from fatal glomerulonephritis in association with a lack of IgG3 and strong reduction in the serum levels of gp70 IC. It has been suggested that IgG3 MAbs of specific yet undefined physicochemical properties might induce glomerular lesions in non-autoimmune mice when produced from transplanted hybridoma cells, regardless of their antigenic specificity (9, 33, 34). One can argue that an extremely high serum concentration of IgG3 would be achieved when hybridoma cells were transplanted, and cryoprecipitating activity of IgG3 molecules induced the observed glomerular lesions. However, it should be noted that hybridoma clones secreting MAbs with strong cryoprecipitating activity were

4125

clearly distinguished from noncryogenerating or less cryogenerating anti-gp70 clones through the screening procedure and, thus eliminated from the present study, because the former actually caused fine granular precipitates throughout the bottoms of culture wells after overnight incubation with target cells at 4°C. In addition, none of the mice transplanted with our anti-gp70 Ab-producing hybridomas, either of IgG3 or another isotype, developed purpuric skin lesions, a typical manifestation of cryoglobulinemia (7, 11). These findings, along with the demonstration of gp70 deposition in the induced glomerular lesions (Fig. 4 and 5) and differences in the pathogenicity of injected anti-gp70 MAbs in strains of mice expressing high and low levels of serum gp70 (Fig. 6), strongly suggest that cognate binding of anti-gp70 MAbs to serum gp70 is mainly responsible for their ability to induce glomerular pathology. Because gp70 was expressed from a cloned cDNA in our experiments, and different isolates of endogenous mouse retroviruses are readily available, these anti-gp70 MAbs might become very useful in identifying Ab-binding epitope structures and in analyzing the ontogenic origins of autoantibody-producing cells. Our preliminary analyses using chimeras between NZB xenotropic viral and SFFV env genes has shown that onehalf of the anti-gp70 autoantibody clones including the highly pathogenic 12H5.1 recognize epitopes located within the Nterminal 150 amino acids of the xenotropic viral gp70. Amino acid sequences are rather homologous in this region between NZB xenotropic virus and SFFV (21, 29, 42). The major difference consists of an insertion of four consecutive amino acids in NZB xenotropic viral gp70, in addition to several scattered amino acid substitutions. Construction of recombinant vaccinia viruses that express gp70 minigenes and use of synthetic oligopeptides may lead to the identification of epitope structures recognized by pathogenic anti-gp70 MAbs in the near future. ACKNOWLEDGMENTS We thank M. Nose for providing SCID mice and scientific advice, M. P. Gorman for reviewing the manuscript, and J. Nishio, H. Shiwaku, E. Kondoh, and Y. Akahoshi for technical assistance. Some of the recombinant vaccinia viruses described in this report were constructed during M. Miyazawa’s stay at the Rocky Mountain Laboratories, Hamilton, Mont., under the financial support of B. Chesebro. This work was supported in part by grants from the Ministries of Education, Science and Culture and of Health and Welfare of Japan and from the Cell Science Foundation. REFERENCES 1. Andrews, B. S., R. A. Eisenberg, A. N. Theofilopoulos, S. Izui, C. B. Wilson, P. J. McConahey, E. D. Murphy, J. B. Roth, and F. J. Dixon. 1978. Spontaneous murine lupus-like syndromes. Clinical and immunopathological manifestations in several strains. J. Exp. Med. 148:1198–1215. 2. Chesebro, B., W. Britt, L. Evans, K. Wehrly, J. Nishio, and M. Cloyd. 1983. Characterization of monoclonal antibodies reactive with murine leukemia viruses: use in analysis of strains of Friend MCF and Friend ecotropic murine leukemia virus. Virology 127:134–148. 3. Chesebro, B., and K. Wehrly. 1976. Studies on the role of the host immune responses in recovery from Friend virus leukemia. I. Antiviral and antileukemia cell antibodies. J. Exp. Med. 143:73–84. 4. Cohen, P. L., and R. A. Eisenberg. 1991. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9:243–269. 5. Elder, J. H., F. C. Jensen, M. L. Bryant, and R. A. Lerner. 1977. Polymorphism of the major envelope glycoprotein (gp70) of murine C-type viruses: virion associated and differentiation antigens encoded by a multi-gene family. Nature 267:23–28. 6. Fijinami, R. S., and M. B. A. Oldstone. 1985. Amino acid homology and immune response between the encephalitogenic site of myelin basic protein and virus: a mechanism of autoimmunity. Science 230:1093–1095. 7. Gyotoku, Y., M. Abdelmoula, F. Spertini, S. Izui, and P.-H. Lambert. 1987. Cryoglobulinemia induced by monoclonal immunoglobulin G rheumatoid factors derived from autoimmune MRL/MpJ-lpr/lpr mice. J. Immunol. 138: 3785–3792.

4126

TABATA ET AL.

8. Hara, I., S. Izui, and F. J. Dixon. 1982. Murine serum glycoprotein gp70 behaves as an acute phase reactant. J. Exp. Med. 155:345–357. 9. Itoh, J., S. Takahashi, M. Ono, T. Yamamoto, M. Nose, and M. Kyogoku. 1994. Nephritogenic antibodies in MRL/lpr lupus mice: molecular characteristics in pathological and genetic aspects. Tohoku J. Exp. Med. 173: 65–174. 10. Iwashiro, M., T. Kondo, T. Shimizu, H. Yamagishi, K. Takahashi, Y. Matsubayashi, T. Masuda, A. Otaka, N. Fujii, A. Ishimoto, M. Miyazawa, M. N. Robertson, B. Chesebro, and K. Kuribayashi. 1993. Multiplicity of virusencoded helper T cell epitopes expressed on FBL-3 tumor cells. J. Virol. 67: 4533–4542. 11. Izui, S., T. Berney, T. Shibata, T. Fulpius, L. Fossati, and R. Merino. 1994. Molecular and cellular basis for pathogenicity of autoantibodies. Tohoku J. Exp. Med. 173:15–30. 12. Izui, S., J. H. Elder, P. J. McConahey, and F. J. Dixon. 1981. Identification of retroviral gp70 and anti-gp70 antibodies involved in circulating immune complexes in NZB ⫻ NZW mice. J. Exp. Med. 153:1151–1160. 13. Madaio, M. P., J. Carlson, J. Cataldo, A. Ucci, P. Migliorini, and O. G. Pankewycz. 1987. Murine monoclonal anti-DNA antibodies bind directly to glomerular antigens and form immune deposits. J. Immunol. 138:2883–2894. 14. Maruyama, N., and C. O. Lindstrom. 1983. H-2-linked regulation of serum gp70 production in mice. Immunogenetics 17:507–521. 15. Maruyama, N., C. O. Lindstrom, H. Sato, and F. J. Dixon. 1983. Serum gp70 production regulated by a gene on murine chromosome 7. Immunogenetics 18:365–371. 16. Miyazawa, M., S. Mori, G. J. Spangrude, J. B. Wolfinbarger, and M. E. Bloom. 1994. Production and characterization of new monoclonal antibodies that distinguish subsets of mink lymphoid cells. Hybridoma 13:107–114. 17. Miyazawa, M., J. Nishio, and B. Chesebro. 1992. Protection against Friend retrovirus-induced leukemia by recombinant vaccinia viruses expressing the gag gene. J. Virol. 66:4497–4507. 18. Miyazawa, M., J. Nishio, M. Kyogoku, and B. Chesebro. 1992. Host genetic control of immune responses to molecularly cloned Friend leukemia virus antigens, p. 177–184. In T. O. Yoshida and J. M. Wilson (ed.), Molecular approaches to the study and treatment of human diseases. Elsevier Science Publishers, B.V., Amsterdam, The Netherlands. 19. Miyazawa, M., M. Nose, M. Kawashima, and M. Kyogoku. 1987. Pathogenesis of arteritis of SL/Ni mice. Possible lytic effect of anti-gp70 antibodies on vascular smooth muscle cells. J. Exp. Med. 166:890–908. 20. Naparstek, Y., and H. P. Plotz. 1993. The role of autoantibodies in autoimmune disease. Annu. Rev. Immunol. 11:79–104. 21. O’Neill, R. R., C. E. Buckler, T. S. Theodore, M. A. Martin, and R. Repaske. 1985. Envelope and long terminal repeat sequences of a cloned infectious NZB xenotropic murine leukemia virus. J. Virol. 53:100–106. 22. Portis, J. L., and F. J. McAtee. 1983. Monoclonal antibodies derived during graft-versus-host reaction. II. Antibodies detect unique determinants common to many MCF viruses. Virology 126:96–105. 23. Portis, J. L., F. J. McAtee, and M. W. Cloyd. 1982. Monoclonal antibodies to xenotropic and MCF murine leukemia viruses derived during the graftversus-host reaction. Virology 118:181–190. 24. Robertson, M. N., M. Miyazawa, S. Mori, B. Caughey, L. H. Evans, S. F. Hayes, and B. Chesebro. 1991. Production of monoclonal antibodies reacting with a denatured form of the Friend murine leukemia virus gp70 envelope protein: use in a focal infectivity assay, immunohistochemical studies, electron microscopy, and Western blotting. J. Virol. Methods 34:255–271. 25. Santiago, M.-L., L. Fossati, C. Jacquet, W. Mu ¨ller, S. Izui, and L. Reininger.

J. VIROL.

26.

27. 28. 29. 30.

31. 32. 33.

34. 35. 36.

37.

38. 39. 40. 41. 42.

1997. Interleukin-4 protects against a genetically linked lupus-like autoimmune syndrome. J. Exp. Med. 185:65–70. Santiago, M.-L., C. Mary, D. Parzy, C. Jacquet, X. Montagutelli, R. M. E. Parkhouse, R. Lemoine, S. Izui, and L. Reininger. 1998. Linkage of a major quantitative trait locus to Yaa gene-induced lupus-like nephritis in (NZW ⫻ C57BL/6)F1 mice. Eur. J. Immunol. 28:4257–4267. Satoh, M., A. Kumar, Y. S. Kanwar, and W. H. Reeves. 1995. Anti-nuclear antibody production and immune-complex glomerulonephritis in BALB/c mice treated with pristane. Proc. Natl. Acad. Sci. USA 92:10934–10938. Satoh, M., and W. H. Reeves. 1994. Induction of lupus-associated autoantibodies in BALB/c mice by intraperitoneal injection of pristane. J. Exp. Med. 180:2341–2346. Shigemoto, K., S. Kubo, Y. Itoh, G. Tate, S. Handa, and N. Maruyama. 1992. Expression and structure of serum gp70 as an acute phase protein in NZB mice. Mol. Immunol. 29:573–582.24. Smith, G. L., B. R. Murphy, and B. Moss. 1983. Construction and characterization of an infectious vaccinia virus recombinant that expresses the influenza hemagglutinin gene and induces resistance to influenza virus infection in hamsters. Proc. Natl. Acad. Sci. USA 80:7155–7159. Stoye, J. P., and J. M. Coffin. 1987. The four classes of endogenous murine leukemia virus: structural relationships and potential for recombination. J. Virol. 61:2659–2669. Sun, L., M. J. Luce, K. Ren, H. Ha, and P. D. Burrows. 1995. Identification of polymorphism in the constant region of IgG3: the missing mouse allotype. Int. Immunol. 7:337–341. Takahashi, S., J. Itoh, M. Nose, M. Ono, T. Yamamoto, and M. Kyogoku. 1993. Cloning and cDNA sequence analysis of nephritogenic monoclonal antibodies derived from an MRL/lpr lupus mouse. Mol. Immunol. 30:177– 182. Takahashi, S., M. Nose, J. Sasaki, T. Yamamoto, and M. Kyogoku. 1991. IgG3 production in MRL/lpr lupus mice is responsible for development of lupus nephritis. J. Immunol. 147:515–519. Theofilopoulos, A. N., and F. J. Dixon. 1985. Murine models of systemic lupus erythematosus. Adv. Immunol. 37:269–390. Tsao, B. P., F. M. Ebling, C. Roman, N. Panosian-Sahakian, K. Calame, and B. H. Hahn. 1990. Structural characteristics of the variable regions of immunoglobulin genes encoding a pathogenic autoantibody in murine lupus. J. Clin. Investig. 85:530–540. Tsurudome, M., A. Yamada, M. Hishiyama, and Y. Ito. 1990. Monoclonal antibodies against the nucleoprotein of mumps virus: their binding characteristics and cross-reactivity with other paramyxoviruses. Acta Virol. 34: 220–227. Vlahakos, D. V., M. H. Foster, S. Adams, M. Katz, A. A. Ucci, K. J. Barrett, S. K. Datta, and M. P. Madaio. 1992. Anti-DNA antibodies form immune deposits at distinct glomerular and vascular sites. Kidney Int. 41:1690–1700. Vogt, A., S. Batsford, and T. Morioka. 1994. Nephritogenic antibodies in lupus nephritis. Tohoku J. Exp. Med. 173:31–41. Vyse, T. J., C. G. Drake, S. J. Rozzo, E. Roper, S. Izui, and B. L. Kotzin. 1996. Genetic linkage of IgG autoantibody production in relation to lupus nephritis in New Zealand hybrid mice. J. Clin. Investig. 98:1762–1772. Vyse, T. J., and B. L. Kotzin. 1998. Genetic susceptibility to systemic lupus erythematosus. Annu. Rev. Immunol. 16:261–292. Wolff, L., E. Scolnick, and S. Ruscetti. 1983. Envelope gene of the Friend spleen focus-forming virus: deletion and insertions in 3⬘ gp70/p15E-encoding region have resulted in unique features in the primary structure of its protein product. Proc. Natl. Acad. Sci. USA 80:4718–4722.